Loading...

Customer Churn Prediction in Therapy Policy in Iran

Khosravi, Mahnaz | 2020

239 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 54696 (44)
  4. University: Sharif University of Technology
  5. Department: Management and Economics
  6. Advisor(s): Aslani, Shirin; Arian, Hamidreza
  7. Abstract:
  8. Using customer relationship management systems has led to the formation of databases containing businesses customer information. Increased competition in the markets, limited resources and high costs of attracting new customers compared to retaining existing customers have made customer retention an inevitable subject for business owners. Meanwhile, insurance companies, which due to their long history have suitable databases of product information and their customers, have resorted to using their databases to manage the relationship with their customers. In this research, while predicting customers churn in health insurance, as one of the most widely used products of domestic insurance companies, the variables affecting it are also identified. For this purpose, the database of one of the domestic insurance companies was examined and while studying the literature, experts in the relevant field were interviewed. Using four machine learning algorithms, including logistic regression, decision tree, support vector machine, and random forest, in python programming language, an attempt was made to identify churned customers with the least error and a model that is interpretable and can help generate managerial insights. In this study, based on the obtained results, it was found that variables such as the percentage of clinical cases with returned referred and no referred, the percentage of SOS cases, the percentage of exit due to death with non-refundable insurance and the ratio of clinical cases to the number of insured are the most important variables in customers churn prediction in the health insurance
  9. Keywords:
  10. Customer Churn Prediction ; Insurance Industry ; Health Insuranee ; Data Mining ; Machine Learning ; Customer Relationship Management

 Digital Object List

 Bookmark

No TOC